Sobetska Olha
Centre Leo Apostel (CLEA), Vrije Universiteit Brussel, Brussels, Belgium.
Front Artif Intell. 2025 Jun 20;8:1579704. doi: 10.3389/frai.2025.1579704. eCollection 2025.
This manuscript explores how human irrationality in decision-making can contribute to artificial intelligence (AI) development, particularly in the domain of creativity. While irrational behavior is typically seen as a cognitive flaw, we argue that certain forms of irrationality, such as those demonstrated by the conjunction fallacy (CF), may represent context-sensitive reasoning that reveals creative problem-solving. Traditional AI research has primarily focused on rational, logic-driven models, overlooking the productive role of non-linear and seemingly illogical human thinking in generating novel insights. Drawing on interdisciplinary insights and recent neuroscientific findings, particularly the interaction of the Default Mode, Executive Control, and Salience Networks, we propose a model that integrates both rational and irrational cognitive dynamics. This framework may inform the design of AI systems that are more adaptive, context-aware, and capable of emulating human-like creativity.
本手稿探讨了人类决策中的非理性如何促进人工智能（AI）的发展，特别是在创造力领域。虽然非理性行为通常被视为一种认知缺陷，但我们认为，某些形式的非理性，如合取谬误（CF）所表现出的那些，可能代表了揭示创造性问题解决方法的情境敏感推理。传统的人工智能研究主要集中在理性的、逻辑驱动的模型上，忽略了非线性且看似不合逻辑的人类思维在产生新颖见解方面的积极作用。借鉴跨学科见解和最近的神经科学发现，特别是默认模式网络、执行控制网络和突显网络的相互作用，我们提出了一个整合了理性和非理性认知动态的模型。这个框架可能为设计更具适应性、情境感知能力且能够模拟类人创造力的人工智能系统提供参考。
